How much should a law firm spend on marketing?
The right number falls out of two things: what a client is worth to your firm, and how well you turn enquiries into instructions. A firm with high-value matters and repeat clients can spend far more, and far more safely, than one with one-off work and no follow-up. Start with your own economics, not an average.
There is no magic percentage. Anyone who gives you one without asking about your firm is guessing.
What the benchmarks actually say #
You will see benchmarks quoted, so here are the real ones. The most-cited global figure is Gartner's: in 2026, companies spend an average of 7.8 per cent of revenue on marketing, barely moved from 7.7 per cent the year before. Worth knowing, and worth reading carefully. That average comes from large enterprises overseas, most turning over more than a billion dollars. It is not a target for a suburban law firm in Perth.
The Australian picture is different. Small firms here typically spend far less, often only two to three per cent of revenue. Read that as a warning if you like, but not as a target either.
Here is the problem with all of these numbers. An average hides the two things that actually decide yours: what a client is worth to your firm, and how much of your marketing leaks before it becomes an instruction. A single conveyance and a commercial client who instructs you for a decade are worth wildly different amounts. Two firms the same size can rationally spend five times apart. Use the benchmarks as a sanity-check, then set your number from your own economics. This is the law-firm version of a general question; here is how much a professional services firm should spend on marketing.
Start with what a client is worth #
You value matters for a living. Value your own firm the same way.
Take one good client. What do they bill you over the years you act for them, and what is your margin on that work? A commercial client worth, say, fifteen thousand a year across several years, at a healthy margin, is a substantial contribution figure.
That number is your ceiling. You can rationally spend a fraction of it to win one client and come out well ahead. The firm that knows this number briefs marketing with confidence. The firm that does not either underspends or freezes.
Then look at what it costs to win one #
Spend only makes sense against a conversion rate.
Of the enquiries you had last month, how many became instructions, and what did the marketing that produced them cost? Most firms cannot say, because nobody built the tracking. That is the first fix, and it costs only attention.
Once you can answer it, the budget nearly sets itself. If it costs a known amount to win a client worth many times that in margin, spending more is not a gamble.
A simple way to set a starting number #
Work forward, not from a percentage.
Decide how many new clients you want this quarter. Multiply by what it costs to win one today. That is your floor. If you do not yet know the cost to win one, start modest, measure everything, and replace the guess with the real number within a month or two.
Then sense-check against the ceiling. If the plan spends a comfortable slice of the margin those clients bring, it is sound. If it spends more than they are worth, the problem is usually conversion, not budget.
Why the budget is wasted if the firm leaks #
The most expensive mistake is spending more into a firm that loses what it catches.
An enquiry that sits over the weekend, no follow-up, no idea which channel produced the client: a bigger budget just buys more of that waste. Fix the leaks and the same spend produces more instructions.
When a client asks AI for a lawyer, is your firm named?
Where AI recommendation fits #
One shift worth budgeting for. Clients now ask AI who is good before they search, and being the named answer costs attention rather than media. It compounds instead of stopping when you stop paying. Ads bring instructions now; AI visibility lowers what you need to spend later. Most firms should do some of both. See how to get your law firm recommended by ChatGPT.
Questions principals ask #
Isn't there a rule of thumb, like a percentage of revenue?
The most-cited global benchmark is about 7.8 per cent of revenue (Gartner, 2026), though that is a large-enterprise average. Australian small firms typically spend far less, often two to three per cent. Both are context, not targets. They ignore what a client is worth to you and how well you convert, which is what should actually set your number.
We have never tracked cost per client. Where do we start?
Tag every enquiry at the source and follow it through to an instruction. Once you can see cost per client by channel, every budget question gets easier.
Ads or being found by AI?
Both, in proportion. Ads produce enquiries now. AI visibility compounds and lowers your cost to win over time.